Sandipan Patra

This paper proposes an advanced modular multilevel converter (MMC)‐based topology specifically developed for medium‐voltage (MV) grid‐connected ultra‐fast electric vehicle (EV) charging applications. A new intra‐phase proportional balancing via d/q injection (IPB‐DQI) strategy is proposed for MMCs, achieving robust submodule (SM) voltage balancing through intra‐phase proportional feedback and implicit generation of circulating currents without added computational complexity. The developed solution maintains stability at 600 V per SM, supporting a 2.4 kV MVDC link using cost‐effective 750 V SiC MOSFET devices based on proven derating practices. Key operational metrics include consistently high‐power quality, minimal harmonic distortion in both current and voltage and stable unity power‐factor operation across varying loads. The enhanced balancing approach significantly mitigates capacitor voltage fluctuations, improves robustness, reduces switching‐device stress and enhances the system's overall efficiency and reliability. Comprehensive validation of the proposed topology was conducted through real‐time performance testing using an OPAL‐RT experimental setup across different case studies. In addition, results from the OPAL‐RT and TI DSP (TMS320F28379D) embedded controller hardware‐in‐the‐loop experimental setup confirm that the proposed balancing method effectively ensures stability, dynamic performance and robust grid integration. This makes the system well‐suited for scalable deployment in industrial‐level fast‐charging infrastructures.

This paper presents a new electric vehicle (EV) integration topology for medium voltage (MV) networks, addressing the challenge of prolonged charging times. The solution leverages the innovative integration of dual active bridge (DAB) converters arranged in an input-series output-parallel (ISOP) configuration. The power distribution infrastructure includes a modular DAB design in an ISOP configuration to increase power scalability. One of the benefits of ISOP configuration is that it highlights a significant reduction in the number of DAB converters required compared to traditional topologies. This streamlined approach not only simplifies the system architecture but also enhances efficiency and decreases the overall cost, making it a compelling option for MV network connection. It further explains the new control approach for ISOP configuration of DAB in support of efficient input voltage sharing (IVS) and output current sharing (OCS) control for controlling EV charging current. The MATLAB-based simulation validates the topology, demonstrating successful control of the 600V input voltage for each DAB connected to the 2.4 kV MVDC link. This system uses inexpensive and readily available SiC 750V MOSFETs, thus improving the practicality of the system. These results reinforce the viability of the topology and control approach, establishing their relevance for extreme-fast EV charging and industrial deployment.

This study introduces an innovative single-phase grid-connected five-level inverter design that features minimized DC link capacitor requirements while enhancing the overall power capacity of the inverter. The reduction in size of DC link capacitance is achieved by using a combination of charge pump circuits and switched capacitors. Existing common ground multilevel inverters suffer from capacitor voltage decay and imbalance, which also limits the power rating of the inverters and increases the size of DC link capacitance. The proposed topology can continuously charge the DC link capacitors, which prevents voltage decay across the capacitors, and the switched capacitor cell avoids the voltage imbalance. The proposed topology significantly reduces the size of DC link capacitors and increases the rated power capacity for common ground five-level configuration. The comprehensive analysis of the proposed topology with recent common ground multilevel topologies is presented. A simulation-based analysis is conducted to assess its performance under dynamic conditions, such as variations in load.

This paper presents a systematic review of recent advancements in V2G cybersecurity, employing the PRISMA (Preferred Reporting Items for Systematic Reviews and Meta-Analyses) framework for detailed searches across three journal databases and included only peer-reviewed studies published between 2020 and 2024 (June). We identified and reviewed 133 V2G cybersecurity studies and found five important insights on existing V2G cybersecurity research. First, most studies (103 of 133) focused on protecting V2G systems against cyber threats, while only seven studies addressed the recovery aspect of the CRML (Cybersecurity Risk Management Lifecycle) function. Second, existing studies have adequately addressed the security of EVs and EVCS (EV charging stations) in V2G systems (112 and 81 of 133 studies, respectively). However, none have focused on the linkage between the behaviour of EV users and the cybersecurity of V2G systems. Third, physical access, control-related vulnerabilities, and user behaviour-related attacks in V2G systems are not addressed significantly. Furthermore, existing studies overlook vulnerabilities and attacks specific to AI and blockchain technologies. Fourth, blockchain, artificial intelligence (AI), encryption, control theory, and optimisation are the main technologies used, and finally, the inclusion of quantum safety within encryption and AI models and AI assurance (AIA) is in a very early stage; only two and one of 133 studies explicitly addressed quantum safety and AIA through explainability. By providing a holistic perspective, this study identifies critical research gaps and outlines future directions for developing robust end-to-end cybersecurity solutions to safeguard V2G systems and support global sustainability goals.

This paper describes a comprehensive energy-related dataset, collected from residential electricity households within an energy community in Ireland, as part of StoreNet project. The data includes local weather parameters and per household power (W) and energy (Wh) components for various aspects such as active power consumption, PV generation, grid import and export, charging and discharging, and the state of charge of energy storage. Additionally, it provides weather data for the location at a 1-minute temporal resolution for the year 2020. The dataset has been validated by comparing measurements that should yield identical results to standard load profiles, with no significant inconsistencies discovered. Validation examples have also been given through the published articles where this dataset has been used to analyse peer-to-peer energy trading benefits for the energy community and decision-making support for aggregators. The dataset aids in understanding patterns in electrical load curves and the duty cycle of energy storage within an energy community. Furthermore, it can assist in comprehending the impact on distribution networks caused by distributed energy storage.

This paper presents a novel nonlinear control approach for the DC-DC bidirectional converter to efficiently operate the battery energy storage systems in DC micro-grids. The control design is based on sliding model back-stepping nonlinear control approach. This paper also proposes a novel equivalent dynamic model for the micro-grid system to verify the effectiveness of the proposed control system. The stability analysis is assessed using the Lyapunov stability theory. The effectiveness and robustness of the proposed method are evaluated through the simulation study and considering the nominal case, parameter uncertainties/variations, and sensor faults scenarios.

This article describes a seamless reconnection topology for power-sharing between distant rural community Microgrids (CMGs), which is based on a double-layered droop controlled (DLDC) frequency restoration scheme. Increased load demand, along with the intermittent nature of renewable energy sources, may result in a power deficit in isolated CMGs. In order to overcome this restriction, the connection of autonomous neighboring CMGs may be a viable alternative to intelligent load shedding. When dealing with active power fluctuation and setting a frequency setpoint becomes difficult in the absence of a grid frequency reference, this DLDC-based approach can resolve the issue. The DLDC adds a self-synchronized feature by parallel shifting of the f-P slope to restore the operating frequency to its nominal value. The difference in frequency enables to shift the voltage axis accordingly through change in V_d and V_q reference of the terminal voltage for appropriate power-sharing. The stability of the proposed controller has been analysed by using a mathematical model considering communication delay in the distributed secondary controller. Finally, the efficacy of the proposed controller is shown via the use of unique Processor-in-the-Loop (PIL) experimental findings, in which OPAL-RT and PLECS RT boxes are used to build both the CMGs and a TI based F28069M microprocessor is utilised as a controller.

The data-driven (DD) is a systematic approach to improve the data and model by deriving/adding features to address the problem identified during the iterative loop of forecasting model development. This article proposes a DD framework for forecasting short-term PV generation and load demand. A framework of three stages with a unique contribution in each stage, such as generalizing data preprocessing steps (stage-1), multivariate feature generation and selection (stage-2), and model hyperparameter tuning (stage-3) for further improvement in forecasting is proposed. It focuses on data as well as forecasting models. The whole process is analyzed using the time-series measured data collected from a real-life demonstration project in Ireland. Data preprocessing is generalized for both generation and demand forecasting under the same framework. The relevant features are selected with the help of the proposed random forest sequential forward feature selection algorithm. Hyperparameters are tuned through tree-structured Parzen estimator algorithm for further improvement. In addition, the performance of the classical autoregressive integrated moving average model is compared with the machine learning-based gate recurrent unit, long short term memory, recurrent neural network, and convolutional neural network models. Results show that the data-driven forecasting model framework systematically improves the model performance. The seasonal variation has also a high impact on the model performances.